Skip to content

Elesdes/ICARUS

Repository files navigation

ICARUS - 5AIBD Annual Project 2023-2024

ICARUS is an open-source student project aimed at developing a system for controlling a non-invasive 3D-printed hand prosthesis through EMG signals using deep reinforcement learning algorithms, thereby exploring the interactions between technology and the human body by leveraging concepts related to human-machine interactions.

Architecture

Architecture

Project Global Setup

Install WSL Distribution

In a Windows terminal, run the following commands to install WSL and set it up:

wsl --install -d Ubuntu
wsl -s Ubuntu

From here, all commands are run in WSL

Install Anaconda

mkdir -p ~/miniconda3
wget https://repo.anaconda.com/miniconda/Miniconda3-latest-Linux-x86_64.sh -O ~/miniconda3/miniconda.sh
bash ~/miniconda3/miniconda.sh -b -u -p ~/miniconda3
rm -rf ~/miniconda3/miniconda.sh
~/miniconda3/bin/conda init bash

After running the above commands, reboot WSL.

Prepare Conda Environment

Before this step, you need to access the project's root directory with WSL.

conda update --all -y
conda create -n ICARUS python=3.11 -y
conda activate ICARUS
yes | pip install -r requirements.txt

If you want to run the API on your local machine, or even run the simulation, please refer to these READMEs :

Work Based On

This project has been influenced by the following works:

  • TactHand (Open-source model of a 3D-printed hand prosthesis)
  • InMoov (An other open-source model of a 3D-printed hand prosthesis)
  • PyBullet (Open-source physics engine for Python)
  • ROS (Open-source framework for robot software development)
  • Gazebo (Open-source robotics simulator integrated with ROS)
  • Pyomyo (Open source code for interfacing with Myo armbands)

We would like to express our gratitude to these authors for their groundbreaking work and contributions to the open-source community.

License

This project is licensed under the Apache License (Version 2.0).

See LICENSE for details.

Citation

If you use or reference this work, please cite it as follows:

@misc{maubras2024icarus,
  title={ICARUS: Deep Reinforcement Learning Control of a Non-Invasive 3D-Printed Prosthesis using EMG Signals},
  author={Juan Maubras, Erwan Duprey, Enzo Leonardo},
  year={2024},
  howpublished={\url{https://github.com/Elesdes/ICARUS}},
  note={Unpublished manuscript},
}

Authors

About

A student project using deep reinforcement learning to control a non-invasive 3D-printed hand prosthesis via EMG signals. Explores human-machine interaction through open-source innovation.

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors